Qdrant

What it is

Qdrant is a vector database written in Rust, purpose-built for storing and querying high-dimensional embeddings at scale. It's the backend you reach for when you need similarity search over vectors — semantic search, RAG pipelines, recommendation engines, image/audio similarity — rather than exact-match lookups. It supports HNSW indexing, payload filtering alongside vector search, quantization for memory efficiency, and both cloud and self-hosted deployment. Falls under "App Definition and Development" in the sense that it's infrastructure your application logic depends on directly, not a platform-level primitive.

Who builds it and why

Qdrant is developed by Qdrant (the company), not a foundation-governed project. It's not part of CNCF, and CNCF maturity is not publicly available because it isn't a CNCF project at all — worth noting since Kubeera tracks it under App Definition and Development, but governance here is commercial, not neutral-foundation. 173 contributors is a healthy number for a database of this scope, but contribution concentration (how much comes from the core team vs. outside) is not publicly available from the stats given. The company's incentive is straightforward: open-source core drives adoption, managed cloud offering monetizes it. That's a normal and sustainable model, but it means roadmap priorities will skew toward what supports the commercial product.

Production readiness signal

32,856 GitHub stars and a release cadence reaching v1.18.2 indicate a mature, actively maintained project. Last commit dated 2026-07-01 (per the data provided) suggests ongoing active development — treat this as current if accurate in your sourcing. Rust as the implementation language is a strong signal for memory safety and performance under load, which matters a lot for a database handling large vector indices in memory. Apache-2.0 licensing is permissive and low-risk for commercial use. What's missing from public signal: no CNCF maturity tier, no founding date, and no visibility into contributor concentration or backing-company financials. If you're doing vendor risk assessment, you'll need to go directly to Qdrant for funding/runway info — not publicly available here.

Who should use this

Teams building RAG applications, semantic search, or recommendation systems who need a dedicated vector store rather than bolting vector search onto a general-purpose database. Good fit if you want self-hosted control (Kubernetes-deployable, Rust performance characteristics) or prefer not to pay for a fully managed vector-only SaaS. Also a solid choice if you need payload filtering combined with vector search in one query — Qdrant's filtering is genuinely good, not an afterthought.

Who should NOT use this

• Teams already on Postgres who need "good enough" vector search. pgvector gets you 80% of the value with zero new infrastructure to operate. Don't add a new database to your stack just because vectors are trendy.
• Small-scale use cases — under a few hundred thousand vectors — where an in-memory library (FAISS, or even a managed embedding API's built-in retrieval) is simpler and cheaper to run.
• Teams wanting a fully managed, zero-ops experience with deep enterprise SLAs and support contracts already in place elsewhere — evaluate Qdrant Cloud specifically, not the OSS project, and compare against Pinecone's managed maturity.
• Organizations requiring CNCF-graduated, foundation-governed infrastructure for compliance or vendor-neutrality reasons — Qdrant is single-vendor governed.

Alternatives

• pgvector — Postgres extension for vector search; the pragmatic choice if you already run Postgres and don't want a new system to operate.
• Pinecone — fully managed, proprietary vector database; less ops overhead, but you're locked into their infrastructure and pricing model.
• Weaviate — open-source vector database with a similar self-hosted/cloud split to Qdrant, GraphQL-first API, worth a direct bake-off if you're comparing OSS options.

Pricing

Qdrant core is fully open source under Apache-2.0 — self-host it for free, no license cost, no feature gating in the OSS version as of this writing. Qdrant Cloud (managed hosting) is a separate paid product with its own pricing tiers; specific pricing details are not publicly available in the data provided — check Qdrant's site directly for current cloud pricing.